I must be new to this concept.  What is the "transposition trick"?

Aric

Hello, Aric.

Using the same samples at the same time can lead to so-called phasing problems, when using two instances of the same instrument. (I'm still waiting for some of the specialists here to comment on when exactly they occur to what extent (see my post above).)

To circumvent them, one can send MIDI data transposed by one half-tone to VI (or whole-tone for the special edition; it is sampled in whole-steps) and then, inside VI, use the pitch bend in the opposite direction to "retune" them to the original (wanted) pitch. That way, another sample is used, and there are no phasing issues.

It is preferable to transpose up and pitch down (because otherwise the vibrato gets sped up) - at least this is what is recommended, as I recall.

In short: We want to hear C!
DAW: D ---to VI instance---> VI [uses D sample]: pitch down a whole step ---to DAW---> sounding C (but which is actually a D sample)

Ideally you would include a MIDI remapping tool in your signal chain (we still wanna hear C):
DAW: C ---transpose MIDI signal up a whole step when sending to VI instance ---> contine as above

The modification is just to ease the workflow, so you can enter the wanted pitches, without having to think about the transposition.

Bendings larger than a whole step are, I think, not recommended, so with the SE this trick will provide two additional "phasing free instances" of an instrument, whereas standard instruments will give you up to four (same proceedings but with half-steps).

-Lukas

Thanks Lukas for the detailed explaination!  How much pitch bend should be used to create a whole step? Should it be set to 127 (whole step up) or (whole step down)?

0 is a whole step down, 127 is a whole step up.

-Lukas

Thanks!

Hello,

after more investigation it struck me that phasing is actually nothing more than (destructive) interference of superpositioned audio-signals. If I sum two identical audio signals with a relative phase of pi (or 180°) they cancel each other out. I always thought there was more to it...

Thus, I come to the conclusion that a "transposition-tricked" instrument is only necessary when stacking patches (or playing "unprocessed" unisono lines) of the same instrument. Any combination of the same instrument (same samples) which underwent individual treatment like panning or (different) MIRx settings (in my case) should therefore not cause any "phasy trouble" - can anyone confirm that? And correct any faults in my argumentation if necessary? (don't hesitate to use technical terms...)

A big thank you and good night.
Lukas

Thanks Lukas for this thread. I am no sound engineer and it was very helpful to read your posts. Thanks,

@LuCsa said:

[...]

Thus, I come to the conclusion that a "transposition-tricked" instrument is only necessary when stacking patches (or playing "unprocessed" unisono lines) of the same instrument. Any combination of the same instrument (same samples) which underwent individual treatment like panning or (different) MIRx settings (in my case) should therefore not cause any "phasy trouble" - can anyone confirm that? And correct any faults in my argumentation if necessary? (don't hesitate to use technical terms...) [...]

Simple panning won't help a lot, you will still hear phasing and cancellation issues. What will help indeed are processings like VI Pro's Humanize feature (which adds variation to pitch and timing) and the timbre-related changes induced by the MIR engine (due to the use of different, position- and rotation-dependent impulse responses as well as different pre-EQ settings used for Character and MIRx).

HTH,

@Dietz said:

Simple panning won't help a lot, you will still hear phasing and cancellation issues. What will help indeed are processings like VI Pro's Humanize feature (which adds variation to pitch and timing) and the timbre-related changes induced by the MIR engine (due to the use of different, position- and rotation-dependent impulse responses as well as different pre-EQ settings used for Character and MIRx).

Hi! Sorry for the late reply.

Yeah, after thinking about it again, panning makes no difference since it just "distributes" the same audio signal differently over the outputs at different volumes... Also, every constant shift of phase might also lead to effective cancellation. At least that's what one can easily try in every DAW by duplicating a track and nudging it against the other.

Can I conclude that, to avoid phasing issues, both (in the extreme case) similar audiosignals have to be processed differently in such a way that it's not just a constant shift in phase? So by EQing, different reverberation, pitch alterations, different timings... anything that would probably minimize the correlation between the former same signals? (I guess, different reverberation alone won't do the trick?)

Kind regards,
Lukas

PS.: It just came to me that we havent covered the main question of this thread... So, default transposition trick for Violin(s) 2, yes or no? 😄 (For now it seems, it's not necessary? Would there be a disadvantage to still just doing it?)

I would be really glad about a definite answer to the main question of the thread, namely "Should I transposition-trick an 'Instrument 2' track as a whole or just for problematic unisono passages with the 'Instrument 1'?" It is actually very close to the question "Will there be disadvantages to transposition tricking that suggest not to do it for the whole track?" (I'm aware that pitching down is generally better than pitching up due to vibrato speed.) I'm using the special editions.

I'm asking again, because I'm about to make some changes in my template.

Thanks!
Lukas